知识卡片

结构图卡 · 6609.e · 生成模型:GPT-6

文档问答区分建库与查询链

理解等级:中级

结构图:

flowchart LR
 D[源文档] --> P[解析和分块]
 P --> E[块嵌入]
 E --> V[检索索引]
 Q[用户问题与可用历史] --> W[独立查询改写]
 W --> R[检索候选证据]
 V --> R
 R --> C[组织问题与上下文]
 C --> G[生成并检查答案]

建库处理决定可检索信息,查询链决定本次使用哪些证据;每个环节都可能产生独立失败。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      Figure 1-5. Workflow of a chatbot application Let’s begin by installing the required libraries. For this setup, we are going to use: LangChain This very popular framework enables building LLM application pipelines. Gradio This library allows you to build LLM-driven user interfaces. Unstructured This is a PDF parsing suite that supports a variety of methods for extracting text from PDFs. Sentence Transformers This library facilitates

    参考来源

    - 位置:《Designing Large Language Model Applications》第01章《大语言模型入门》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Designing Large Language Model Applications/_epub-src/OEBPS/ch01.html) - 结论依据:摘录中关于文档问答区分建库与查询链的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:Figure 1-5. Workflow of a chatbot application Let’s begin by installing the required libraries. For this setup, we are going to use: LangChain This very popular framework enables building LLM application pipelines. Gradio This library allows you to build LLM-driven user interfaces. Unstructured This is a PDF parsing suite that supports a variety of methods for extracting text from PDFs. Sentence Transformers This library facilitates